Seafloor sediments have a significant role in the planning and development of coastal areas, especially port areas. Acoustic technology developing today, especially multifrequency Multibeam Echosounder (MBES), is expected to measure seafloor sediment and detect the type and distribution of seafloor sediments. The question posed in this study is how to improve the accuracy of sediment classification using multifrequency MBES. This study uses deep neural networks to classify seafloor sediments in the study area with input bathymetric and bathymetric differences and 74 in situ sediment samples (silt, clayey sand, silty sand, and sandy silt). Sediment classification results show that clayey sand dominates the sediment distribution in the Central and Eastern regions. On the other hand, sandy silt predominates in the western area (harbor pond). Classification of seafloor sediments in the study area has an accuracy of 41.9% (average) and a kappa coefficient of 21.9% (fair). The implication of the study is that bathymetric and bathymetric differences from multifrequency MBES produce a low sediment classification accuracy value of below 50%. Therefore, it needs to be re-evaluated in relation to bathymetric and bathymetric differences and the amount and distribution of sediment sample data needed to improve its accuracy.

Original languageEnglish
Pages (from-to)9-17
Number of pages9
JournalInternational Journal of GEOMATE
Issue number115
Publication statusPublished - 2024


  • Acoustic
  • Coastal areas
  • Deep neural network
  • Kappa coefficient
  • Overall accuracy


Dive into the research topics of 'BATHYMETRY AND BATHYMETRIC DIFFERENCES MBES MULTIFREQUENCY FOR SEAFLOOR SEDIMENT MAPPING'. Together they form a unique fingerprint.

Cite this